JP2009081636A - Image recording apparatus and photographing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image recording apparatus and photographing method which focuses on a plurality of human faces located away from each other in the depth direction. <P>SOLUTION: In one embodiment of the image recording apparatus, an object image is temporarily loaded as an image for reference via an imaging means, and focus adjustment is performed when a release operation is detected, and under the focus-adjusted state, the object image is loaded and recorded as an image for recording via the imaging means. The image recording apparatus includes: a face region specifying means for specifying at least two face regions in the image for reference; and a focus changing continuous shooting means which sequentially makes focus adjustment for the plurality of face regions specified by the face region specifying means, one by one, when one release operation is detected, and takes the focus-adjusted object images one by one as images for recording via the imaging means. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image recording apparatus and an image capturing method having an autofocus (AF) function and a high-speed continuous shooting function, represented by an electronic camera.

  Conventionally, an electronic camera is generally provided with an autofocus (AF) function. In an electronic camera equipped with an AF function, a distance measurement area of a subject image captured from an imaging means equipped with a CCD or the like, such as a partial area in the center of a finder image, is measured using a contrast detection method. Automatically adjust the focus distance. In addition to the central partial area in the finder image, the distance measurement area may be determined by selecting one from a plurality of positions on the finder image by manual operation.

  Some recent electronic cameras have a face detection function. In this electronic camera, for example, in a shooting preparation process started by half-pressing the release button, face recognition processing is performed on the entire area of the subject image (the thinned subject image at this stage) captured from the imaging means. . If there is an area recognized as a face, an area corresponding to that area is selected from the divided screens obtained by dividing the subject image, and then AF processing is performed. (For example, refer patent document 1 etc.).

The face recognition process is performed by a known method. An example is as follows.
In the first method, the outline of the skin color area is extracted from the subject image based on the color, and the degree of matching with the face outline template is measured to recognize that it is a face. As a second method, recognition is performed by obtaining a candidate eye region and measuring the degree of matching with the eye template. As a third method, recognition is performed based on a feature amount obtained by two-dimensional Fourier transform of a face candidate region and a feature amount obtained by two-dimensional Fourier transform of a face template image.

  The AF process is performed using a contrast detection method. In AF processing using the contrast detection method, the focus lens is adjusted to a position where the contrast of the image in the determined distance measurement area is maximized while moving the focus lens back and forth.

JP 2006-227080 A

When shooting multiple people with an electronic camera, the electronic camera is usually placed so that the person is positioned in the center of the viewfinder image, and the person in the center is focused by AF processing. Take a complete picture in focus.
When shooting an area with a face detected by the face recognition function as a distance measurement area, select one of the detected face areas and focus on it as the distance measurement area. Shoot.

In any case, in the conventional photographing method, the entire subject needs to be positioned within the focusing range of the person who is focused. If it is out of focus, it will be out of focus. For this reason, if you leave multiple people's faces in memory, such as a group photo, if you are out of focus and you try to shoot again, or if you still get out of focus, arrange the people so that they do not have depth. Need to change.

As described above, the conventional electronic camera is not suitable for photographing a composition in which a plurality of persons are located apart in the depth direction.
SUMMARY An advantage of some aspects of the invention is that it provides an image recording apparatus and a photographing method capable of photographing while focusing on the faces of a plurality of persons located apart in the depth direction.

  One aspect of the image recording apparatus of the present invention is that the subject image is temporarily captured as a reference image through the imaging unit, and when the release operation is detected, the focus adjustment is performed and the subject image under the focus adjustment is captured by the imaging unit. An image recording apparatus that captures and records as a recording image through a face area specifying unit that specifies at least two face areas in the reference image, and detecting the face by detecting a single release operation. Focus adjustment in a plurality of face areas specified by the area specifying means for each face area sequentially, and focus variation continuous shooting means for capturing a subject image under each focus adjustment as a recording image through the imaging means; , So that it has.

  The focus variation continuous shooting means groups a plurality of face areas specified by the face area specifying means according to a depth distance by detecting a single release operation, and represents representative face areas in each group. It is also possible to sequentially adjust the focus and to capture a subject image under each focus adjustment as a recording image through the imaging means.

In addition, setting means for setting the number of times of continuous shooting is further provided, and the focus variation continuous shooting means groups the plurality of face areas specified by the face area specifying means for the set number of times of continuous shooting according to the distance of the depth. You may comprise as follows.

  The focus variation continuous shooting means is configured to continuously open and close the shutter and detect the focus according to a predetermined area order for each period during which the shutter is closed by detecting a single release operation. May be.

Further, image synthesizing means for synthesizing the focusing area of each recording image with the reference image using one of the plurality of recording images captured under the respective focus adjustments as a reference image. It is preferable to constitute so that
Further, it is preferable to have a composition information embedding unit that embeds the focus position of each recording image captured under each focus adjustment in the recording image as information at the time of image synthesis.

One aspect of the photographing method of the present invention is to automatically switch the autofocus area to each face area recognized as a face by face recognition processing every time shooting is performed in high-speed continuous shooting. .
It should be noted that the image of the partial area that has become the autofocus area is extracted from a plurality of recording images generated by shooting each face area as an autofocus area in a predetermined order and synthesized. It is preferable to do.

  According to the present invention, it is possible to perform shooting by focusing on the faces of a plurality of persons located away in the depth direction with a single release operation.

Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings.
(First embodiment)
FIG. 1 is a block diagram of an electronic camera according to a first embodiment of the image recording apparatus of the present invention.

The electronic camera 1 includes a focus lens 2, a zoom lens 3, a lens driving block 4, a diaphragm mechanism 5, a CCD 6, a vertical driver 7, a TG (timing generator) 8, a CDS / AGC / AD unit circuit 9, a signal processing unit 10, A CPU 11, a DRAM 12, a ROM 13, a flash memory 14, a key input unit 15, and an image display unit 16 are provided.
The focus lens 2 and the zoom lens 3 are each composed of a lens group (not shown). The lens drive block 4 includes a focus motor 40a that drives the focus lens 2 in the optical axis direction, a zoom motor 40b that drives the zoom lens 3 in the optical axis direction, and a focus motor 40a and a zoom motor 40b according to control signals sent from the CPU 11. Are configured by motor drivers 41a and 41b.
The diaphragm mechanism 5 adjusts a diaphragm (not shown) by a control signal from the CPU 11.

  The CCD 6 has an image pickup surface in which a large number of light receiving elements are two-dimensionally arranged. The CCD 6 photoelectrically converts a subject image projected on the image pickup surface through a focus lens 2, a zoom lens 3, and a diaphragm (not shown), and is vertically The driver 7 drives to output an image signal for one frame every predetermined period. The operation timings of the vertical driver 7 and the CDS / AGC / AD unit circuit 9 are appropriately adjusted by the CPU 11 controlling the TG 8 that generates a timing signal having a predetermined frequency.

  The CDS / AGC / AD unit circuit 9 is a CDS (Correlated Double Sampling) circuit that holds the imaged signal output from the CCD 6 by correlated double sampling, and AGC (Automatic Gain) that performs automatic gain adjustment of the imaged signal after the sampling. Control) circuit and an A / D converter that converts the analog image pickup signal after the automatic gain adjustment into a digital signal. The imaging signal from the CCD 6 that has passed through the CDS / AGC / AD unit circuit 9 is sent to the signal processing unit 10 as a digital signal for each frame.

  The signal processing unit 10 includes a color process circuit, a DMA controller, and the like, and a luminance color difference signal (YCbCr) as a subject image for one frame obtained by performing color process processing on the output from the CDS / AGC / AD unit circuit 9. ) Is transferred to a specified area of the DRAM 12 by DMA (direct memory access) and expanded, and the output from the CDS / AGC / AD unit circuit 9 is subjected to color process processing, AF detection signal, AE detection signal, and AWB detection The signal is output to the CPU 11. In addition, a circuit for performing processing such as defective pixel correction, gamma correction, interpolation, conversion, edge enhancement, and the like are also provided, and these processing is also appropriately performed on the output from the CDS / AGC / AD unit circuit 9. . Furthermore, a JPEG compression / decompression circuit is also provided.

  The CPU 11 includes a calculation unit, a control unit, and the like, and implements control of various units of the electronic camera 1 and various photographing functions by appropriately reading and executing predetermined programs and data recorded in the ROM 13. Although not shown, the CPU 11 further includes a clock circuit, and realizes functions of date and time counting and a timer.

  The ROM 13 stores a program necessary for the CPU 11 to control each unit, a program for realizing various functions, and data necessary for each. Here, programs and various data necessary for the face recognition AF continuous shooting process described later in the CPU 11 are also recorded.

  The DRAM 12 is a means for temporarily storing data, a buffer area for holding the subject image (luminance color difference signal) generated by the signal processing unit 10 described above, and a continuous image described later calculated by the CPU 11 in the high-speed continuous shooting setting. A buffer area for holding information and a later-described reference image and composite image stored in the flash memory 14 and a work area for the CPU 11 are allocated.

The flash memory 14 is a detachable recording medium, and is provided for storing a recording image such as a reference image and a composite image described later.
The image display unit 16 includes a color LCD and its driving circuit. When the recording mode is selected, the image display unit 16 displays a subject image (a thinned subject image or a recording image as a reference image) developed in the DRAM 12. Displays moving images or still images as images. When the playback mode is selected, the recorded image read from the flash memory 14 and expanded is displayed as a finder image. Further, file list information in the flash memory is displayed as appropriate.

  The key input unit 15 includes a mode switching button used for selecting a recording mode and a playback mode, a release button for instructing start and end of single shooting or continuous shooting (continuous shooting), an aperture, a shutter speed, and continuous shooting. Includes various operation buttons such as a setting button for setting the number of sheets, a display / non-display switching button for selecting display / non-display of image information on the LCD, and a zoom button used for changing the zoom magnification. Then, an operation signal corresponding to the button operation is sent to the CPU 11. Further, the operation button has a so-called half shutter function, and is configured to be able to perform a half press operation for shooting advance notice and a full press operation for actual shooting instruction.

FIG. 2 is a block diagram of the color process circuit of the signal processing unit 10.
The color process circuit 100 includes a red component integrator 1001, a green component integrator 1002, a blue component integrator 1003, an AE integrator 1004, an AF filter 1005, various counters 1006, an AWB circuit 1007, and a signal synthesis circuit 1008. ing.

In the color process circuit 100, an AF detection signal, an AE detection signal, an AWB detection signal, and a luminance color difference signal (YCbCr) for the subject image for one frame output from the A / D converter 90 of the CDS / AGC / AD unit circuit 9 are used. ) Generation processing.
The AF filter 1005 applies a high-pass filter (HPF) to the input signal as the subject image to extract a high-frequency signal and outputs the integrated value to the CPU 11 as an AF detection signal. The AE integrator 1004 integrates the input signal and outputs the integrated value to the CPU 11 as an AE detection signal.

Further, the AWB circuit 1007 outputs the integrated value of each component integrated by the red component integrator 1001, the green component integrator 1002, and the blue component integrator 1003 to the CPU 11 as an AWB detection signal.
The counter group 1006 includes an AF window size control counter. When the CPU 11 designates an area to be detected for the counter group 1006, the AF filter 1005 integrates the pixel data of the designated area based on the count value of the pixels counted by the AF window size control counter, and The integral value of the specified area is output.

An example of continuous shooting processing performed under this configuration will be shown below.
Example 1
FIG. 3 is a memory configuration diagram of a buffer area for holding continuous shot image information in the DRAM 12.

In this example, the continuous shot image information calculated from the subject image in the CPU 11 is referred to as “face position information” and “focus lens adjustment position information”, and “serial number information”, “face position information”, “ A storage area for “focus lens adjustment position information” is provided.
“Sequence number information” contains a unique number indicating the shooting order, “Face position information” contains information indicating the position of one of the face areas in the subject image, and “Focus” “Lens adjustment position information” contains information indicating the adjustment position of the focus lens when focus adjustment is performed in one face area of “face position information”.

FIG. 4 is an explanatory diagram of the position information of the face and the adjustment position information of the focus lens.
FIG. 4A shows an example of a reference image temporarily stored in an image buffer area in the DRAM 12 expressed in two-dimensional display coordinates, and surrounds a two-dimensional coordinate axis (XY axis) and a face area. An oval window frame is superimposed on the image. In this example, since 10 persons are included in the image diagram, when the face recognition process is performed in the CPU 11, each face area for 10 persons is displayed as position information on a two-dimensional display coordinate (here, a circular window). Center position information). The “face position information” in the DRAM 12 stores center position information (Xn, Yn) (where n is an integer) of the circular window.

  FIG. 4B shows the composition of the camera and the subject from obliquely above when the reference image of the image of FIG. 4A is captured through the imaging means. As shown in FIG. 4 (b), the distances of 10 people from the electronic camera are different, and it can be seen that there is a large difference in distance between the person who is most in trouble and the person who is farthest away. . In such a composition, when all the 10 people do not fit within the depth of field of the camera when focusing on one person and shooting the whole image, the person who is positioned away from the person who is in focus or in the back An image in which the face is out of focus is generated. In such a case, in place of the continuous shooting setting, images for the number of persons who are focused on each person (particularly the face) are taken.

  Focusing on each person is performed under the control of the CPU 11. Specifically, the display coordinate position of a person's face area out of the face areas for 10 persons extracted by the face recognition process is set in the AF window control counter, and the person's face area output from the AF filter While the AF detection signal is monitored, the moving direction and the moving amount from the reference position of the focus lens when the AF detection signal takes the maximum value by the contrast detection method are measured. Then, the positive and negative movement amounts are stored in “focus lens movement position information” in the DRAM 12.

Below, the process of CPU11 regarding continuous shooting is demonstrated.
Each process is performed by the CPU 11 appropriately reading and executing various programs recorded in the ROM 13.

FIG. 5 is a processing flow of the CPU 11 during continuous shooting.
This processing is started by detecting the half-pressed state of the release button from the user when the electronic camera is set to the recording mode, and further, automatic exposure correction (AE) and auto white balance (AWB). ) Shall be performed in a fixed manner.

First, it is determined whether the face recognition AF function provided in the electronic camera is on or off (S1). If it is in the OFF state, the process proceeds to a conventional shooting process flow, and single shooting or continuous shooting is performed.
On the other hand, if it is ON, then it is determined whether the face recognition AF function provided in the electronic camera and the high-speed continuous shooting linkage function are ON or OFF ( S2). If the camera is in the OFF state, the process proceeds to the flow of the conventional face recognition AF shooting process and shooting is performed.

On the other hand, if it is in the ON state, the face recognition AF continuous shooting process from the subsequent step S3 is executed.
In the face recognition AF continuous shooting process, first, the number of face regions in a subject image (here, a reference image) is acquired (S3). In this processing, for example, face recognition processing is performed on the entire area of the subject image (here, the reference image) temporarily stored in the DRAM 12 to acquire position information for each face area, and each time the position information is acquired. The number of face areas is acquired by counting up the number of face areas from zero. The acquired position information is stored in the “face position information” while assigning a serial number to the “serial number information” in the continuous image information buffer area in the DRAM 12. In this way, two or more face areas are specified. Note that the face recognition process itself is performed by a known method. Here, a method of recognizing that a face is extracted by extracting the outline of a skin color region from the subject image based on the color and measuring the degree of matching with the face outline template (for details, see Japanese Patent Application Laid-Open No. 2006-227080). If there is an area recognized as a face, two-dimensional coordinate information such as the center position of the area is acquired.

Next, the number of face areas acquired in step S3 is given to a parameter for designating the number of continuous shots (S4).
Next, the face recognition AF function and the high-speed continuous shooting function are linked to perform recording of the subject image for the number of times given to the parameter (S5).
This process is performed, for example, as follows after detecting the release button full-press state from the user only once.

  First, position information whose serial number information is “0” is acquired from the continuous image information buffer area, the area is designated to the AF window size control counter of the signal processing unit 10, and AF detection output from the AF filter is performed. Monitor the signal. Then, the area is focused by a contrast detection method. Specifically, while moving the focus lens forward or backward along the optical axis from the current position, the position of the focus lens when the AF detection signal takes the maximum value is searched, and the focus lens is obtained at the position where the maximum value is taken. Is fixed (face recognition AF processing). The recording image (luminance color difference signal) for one frame is temporarily stored in the DRAM 12 through the processing of the CCD 6, the CDS / AGC / AD unit circuit 9, and the signal processing unit 10 by controlling the TG 8 and the signal processing unit 10. Since the CPU 11 knows which number of face area is currently selected, the CPU 11 subsequently reads the position information of the corresponding serial number in the continuous image information buffer area, and reads the position information (focus position). ) In the subject image, for example, in the form of EXIF (EXchangeable Image File format) as continuous image composition information (composition information when combining continuous images). Then, a recording image (embedded image) including the embedded information is JPEG compressed by the signal processing unit 10, and a plurality of patterns are continuously captured by changing the file name of the serial number, for example, the scene, to the embedded compressed data. Assuming the case, a sequential shooting identification symbol in one scene is attached to the head and a file name assigned with a serial number is assigned and stored in the flash memory 14.

In this way, the shooting process is performed for each time in the continuous shooting, and the same shooting process is performed for all of the remaining serial number information in order from the subsequent serial number information “1” (high-speed continuous shooting process). .
In other words, in this process, the shutter (here, the electronic shutter) is opened and closed continuously upon detection of a single release operation, and is automatically focused in a predetermined area order for each period during which the shutter is closed. Adjustment is performed (automatic focus area is changed automatically to change focus and continuous shooting). Then, the embedded compressed data corresponding to the number of times given to the parameter is stored in the flash memory 14.

  It is to be noted that the sum of positive and negative numerical information indicating the back and forth movement instructed to the focus lens motor driver during the face recognition process is counted, and the sum information is added to the “focus lens” of the serial number information “1” in the continuous image information buffer area. Is stored in the “adjustment position information”, and this is performed on all face areas before the recording image is captured. High-speed continuous shooting may be performed by sequentially adjusting the position of the focus lens.

FIG. 6 is a format example of an embedded image.
Here, the embedded image is composed of a “header part” for writing setting information at the time of image recording and a “main image part” including a recording image, and a writing area for continuous image synthesis information is included in the header part. Provided. The coordinates of the focus position on the display coordinates are input to the continuous image composition information writing area.

Next, a process for synthesizing the format image stored in the flash memory 14 by the process shown in FIG. 5 will be described.
FIG. 7 is a processing flow of the CPU 11 at the time of image composition.
Here, the processing is started when the user selects one embedded compressed data from the key input unit 15 in a state where saved information (for example, file name list information) in the flash memory 14 is displayed on the LCD in the playback mode. Is done.

  First, when it is detected that one embedded compressed data is selected by the user (SS1), the serial number file name of the data selected by the user among the embedded compressed data in the flash memory 14 is acquired, and the serial number is acquired. A list of file names is displayed on the LCD (SS2). The user selects, from the list information, the embedded compressed data to be used for synthesizing the focus adjustment area with respect to the initially selected embedded compressed data (here, this main image is the reference image). As a selection method, an operation for collectively specifying all items shown in the list information and an operation for individually specifying each item are registered in the key input unit 15 in advance.

  Subsequently, when a designation signal for embedding compressed data used for synthesizing the focus adjustment area is detected by the user, the designated embedding compressed data is read from the flash memory 14 and decompressed, and the face designated in the header portion. A certain range (focusing area) is copied from the position of (2) and synthesized with the reference image (SS3).

  FIG. 8 is an image diagram when the main image of the embedded image is represented by two-dimensional display coordinates. In the image diagram, a region where focus adjustment has been performed is indicated by a rectangle. A specific example of the process of step SS3 will be described below with reference to this figure.

  First, one embedded compressed data selected first in step SS1 is read from the flash memory 14 and decompressed, and the main image (reference image) therein is expanded in a designated area of the DRAM 12 (FIG. 8A). ). Similarly, any one of the designated embedded compression data for synthesis that follows is read out from the flash memory 14 and decompressed, and the main image is expanded in the designated area of the DRAM 12 (FIG. 8 ( b)). Then, an area of a certain range, for example, the face outline, is covered with the face position information (focus adjustment position) included as continuous image composition information in the header portion after one of the images for composition is expanded. The region (rectangular region 800 in FIG. 8B) is copied from the embedding image for composition, and is combined with the corresponding position on the reference image (rectangular region 801 in FIG. 8C). Further, the composition processing is repeated one by one for the remaining designated embedded compressed data for synthesis in the same manner as described above, and the designated number of focus adjustment area images for one reference image (FIG. 8). The solid rectangular areas 801 to 809 in (d) are synthesized (FIG. 8D). Note that a broken-line rectangular area 810 in FIG. 8D represents a focus adjustment area originally included in the reference image. When the synthesis processing for the specified number of sheets is completed, the synthesized reference image is JPEG compressed by the signal processing unit 10, a new file name is given to the compressed data, and the new image is stored in the flash memory 14.

  As described above, in the first embodiment, even if the subject is located away in the depth direction, it is possible to perform shooting while focusing on the faces of all persons with a single release operation.

(Example 2)
In the second embodiment, a process for specifying the number of continuous shots and combining the number of continuous shots will be described.
FIG. 9 is a memory configuration diagram of a buffer area for holding continuous shot image information in the DRAM 12.
In this example, a storage area for “group number” is further provided in the memory configuration of the first embodiment.
This “group number” contains a group number to which each of a plurality of face regions in the subject image belongs when divided into several groups.

FIG. 10 is an explanatory diagram of how to determine the group number.
FIG. 4 shows the composition of the camera and subject when the reference image of the image in FIG. 4A is captured through the imaging means, as shown in FIG.
In the second embodiment, each person is divided into a plurality of groups according to the distance from the camera, as shown in FIG. Or, the number of images in the number of groups that are in focus on the shaded area or the like are taken.

  The number of groups may be set arbitrarily, but here it is set to the number of continuous shots set for the electronic camera. The area of each group may be specified as appropriate, but in this case, the foremost person and the farthest person are always photographed, and the remaining two shots are subtracted from the number of consecutive shots. The remaining area is equally divided from the camera as the starting point, and each person belongs to the line segment nearest to the standing position. For example, if group numbers 1 to 5 are assigned in order from the front to the depth, the person on the foremost line segment is group number 1, the two in the shaded area in the next row are group number 2, and the next back row Three people in the shaded area are grouped as group number 3, one person on the back line segment is group number 4, and three in the last shaded area are grouped as group number 3. .

Below, the process of CPU11 regarding continuous shooting is demonstrated.
Each process is performed by the CPU 11 appropriately reading and executing various programs recorded in the ROM 13.

FIG. 11 is a processing flow of the CPU 11 during continuous shooting.
This processing is started by detecting the half-pressed state of the release button from the user when the electronic camera is set to the recording mode, and further, automatic exposure correction (AE) and auto white balance (AWB). ) Shall be performed in a fixed manner.
First, it is determined whether the face recognition AF function provided in the electronic camera is in an ON state or an OFF state (SA1). If it is in the OFF state, the process proceeds to a conventional shooting process flow, and single shooting or continuous shooting is performed.

On the other hand, if it is ON, then it is determined whether the face recognition AF function provided in the electronic camera and the high-speed continuous shooting linkage function are ON or OFF ( SA2). If the camera is in the OFF state, the process proceeds to the flow of the conventional face recognition AF shooting process and shooting is performed.
On the other hand, if it is in the ON state, the face recognition AF continuous shooting process from the subsequent step SA3 is executed.

In the face recognition AF continuous shooting process, first of all, continuous shooting or a screen for requesting input of the number of continuous shooting is displayed on the LCD, for example, and the user has designated input from the setting button of the key input unit. The number of times is given to a parameter for designating the number of times of continuous shooting (SA3).
Subsequently, the position in the depth direction of the person in the subject image (here, the reference image) is measured to determine the distance between the foremost person and the farthest person (the distance between the faces in the depth direction) (SA4). .
Further, the focus adjustment position (that is, the position of each line segment in FIG. 10) corresponding to the number of photographing is obtained by equally dividing the face interval in the depth direction obtained in step SA4 by the number of continuous shots given in step SA3. (SA5).
Then, the face recognition AF function and the high-speed continuous shooting function are interlocked to record and shoot the subject image while sequentially adjusting the focus on each line segment (SA6).

The processing of step SA4 to step SA6 is performed by detecting the fully pressed state of the release button from the user. For example:
First, a known face recognition process is performed on the entire area of the subject image (here, the reference image) temporarily stored in the DRAM 12 to obtain position information for each face area, and the continuous shot image information buffer area in the DRAM 12 is acquired. The position information is stored in the “face position information” while assigning serial numbers (here, 1, 2,..., 10) to the “serial number information”.

  Next, the position information “X1, Y1” whose serial number information is “1” is acquired from the continuous image information buffer area, the area is designated to the AF window size control counter of the signal processing unit 10, and AF The AF detection signal output from the filter is monitored. Then, the area is focused by a contrast detection method. Specifically, the position of the focus lens when the AF detection signal takes the maximum value is searched while moving the focus lens forward or backward along the optical axis from the current position. Then, the sum of positive and negative numerical information indicating the back and forth movement instructed to the motor driver of the focus lens during the face recognition process is counted, and the total information (for example, numerical information having a positive or negative sign) is counted in the continuous shot image information buffer area. It is stored in “focus lens adjustment position information” corresponding to the serial number information “1”. Subsequently, the same processing is sequentially performed on all of the serial number information following “1”, and all the “focus lens adjustment position information” to which the serial number information is assigned in the continuous image information buffer area is filled. .

  Next, “adjustment position information of the focus lens” is extracted. Then, the difference between the minimum value and the maximum value is calculated, and the difference is divided by the number obtained by subtracting 2 from the number of continuous shots (the number of line segments at the intermediate position), and this value is added to the minimum value. This is repeated as many times as the number of line segments at the intermediate position, and group information including the sum of each time, the minimum value, and the maximum value is generated. Then, the adjustment position information of the focus lens and the group information are compared, and each adjustment position information of the focus lens is associated with the group information having the smallest difference from the group information (determination of the representative face area). For example, if five group values are set in the group information, the corresponding “focus lens adjustment position information” value and “group information” value are sequentially set from the serial number “1”. The difference is calculated and the value of the group having the smallest difference is associated with the serial number being processed (initially “1”).

Next, the difference between the value of “focus lens adjustment position information” in which the same value is set in “group information” and the group value is calculated, and the difference becomes the smallest in “focus lens adjustment position information”. The value is set as the value of “focus lens adjustment position information” of each serial number corresponding to the group value, the “group information” in the DRAM 12 is set, and the value of “focus lens adjustment position information” is set. Update.
Then, all the updated values of the “focus lens adjustment position information” are acquired to eliminate duplication, and then high-speed continuous shooting is performed by repeating the shooting while sequentially adjusting the position of the focus lens.

Processing for position frames during high-speed continuous shooting is performed as follows.
The recording image (luminance color difference signal) for one frame is temporarily stored in the DRAM 12 through the processing of the CCD 6, the CDS / AGC / AD unit circuit 9, and the signal processing unit 10 by controlling the TG 8 and the signal processing unit 10. Since the CPU 11 knows which face area is currently selected, the CPU 11 subsequently reads the value of the “group information” of the corresponding serial number in the continuous image information buffer area, and reads this value as “group”. All the values of “face position information” included in “information” are read from the continuous-shot image information buffer area, and such information is stored in the subject image in an EXIF (EXchangeable Image File format) format, for example, as shown in FIG. And embedded as continuous image composition information (composition information when combining continuous images). Then, a recording image (embedded image) including the embedded information is JPEG compressed by the signal processing unit 10, and a plurality of patterns are continuously captured by changing the file name of the serial number, for example, the scene, to the embedded compressed data. Assuming the case, a sequential shooting identification symbol in one scene is attached to the head and a file name assigned with a serial number is assigned and stored in the flash memory 14.

Next, a process for synthesizing continuously shot images stored in the flash memory 14 by the process shown in FIG. 11 will be described.
FIG. 12 is a processing flow of the CPU 11 at the time of image composition in the second embodiment.
Here, the processing is started when the user selects one embedded compressed data from the key input unit 15 in a state where saved information (for example, file name list information) in the flash memory 14 is displayed on the LCD in the playback mode. Is done.

First, when it is detected that one embedded compressed data is selected by the user (SSA1), the serial number file name of the data selected by the user among the embedded compressed data in the flash memory 14 is acquired, and the serial number is acquired. A list of file names is displayed on the LCD (SSA2). The user selects, from the list information, the embedded compressed data to be used for synthesizing the focus adjustment area with respect to the initially selected embedded compressed data (here, this main image is the reference image). As a selection method, an operation for collectively specifying all items shown in the list information and an operation for individually specifying each item are registered in the key input unit 15 in advance.
Subsequently, when a designation signal for embedding compressed data used for synthesizing the focus adjustment area is detected by the user, the designated embedding compressed data is read from the flash memory 14 and decompressed, and the one specified in the header portion is read out. A certain range is copied at the face position or the face positions at a plurality of places and synthesized with the reference image (SSA3). When the synthesis processing for the specified number of sheets is completed, the synthesized reference image is JPEG compressed by the signal processing unit 10, a new file name is given to the compressed data, and the new image is stored in the flash memory 14.

  As described above, in the second embodiment, even if the subject is located away from the subject in the depth direction, it is possible to perform shooting with a single release operation by greatly suppressing the defocusing of the person's face.

  In the above embodiment and each example, only an example of the image recording apparatus and the photographing method of the present invention is shown. Therefore, the present invention is not limited to the above configuration and processing procedure, and may take other forms as appropriate. For example, the CPU 11 may be configured to generate AF detection signals, AE detection signals, and AWB detection signals generated by hardware in the color process processing of the signal processing unit 10 by software processing. Alternatively, a part of the software processing in the CPU 11 may be realized by a hardware configuration.

1 is a block diagram of an electronic camera according to a first embodiment of an image recording apparatus of the present invention. 2 is a block diagram of a color process circuit of a signal processing unit 10. FIG. 3 is a memory configuration diagram of a buffer area for holding continuous shot image information according to Embodiment 1. FIG. It is explanatory drawing of the positional information on a face, and the adjustment position information of the said focus lens. 6 is a processing flow of the CPU 11 during continuous shooting according to the first embodiment. It is an example of a format of an embedded image. 3 is a processing flow of the CPU 11 at the time of image composition according to the first embodiment. It is an image figure when the main image of an embedded image is represented by the two-dimensional display coordinate. FIG. 10 is a memory configuration diagram of a buffer area for holding continuous shot image information according to the second embodiment. It is explanatory drawing of how to determine a group number. 6 is a processing flow of the CPU 11 during continuous shooting according to the second embodiment. 12 is a processing flow of the CPU 11 at the time of image composition according to the second embodiment.

Explanation of symbols

1 Electronic Camera 2 Focus Lens 3 Zoom Lens 4 Lens Drive Block 5 Aperture Mechanism 6 CCD
7 Vertical driver 8 TG
9 CDS / AGC / AD unit circuit 10 Signal processor 11 CPU
12 DRAM
13 ROM
14 Flash memory 15 Key input unit 16 Image display unit

Claims (8)

An image recording apparatus that temporarily captures a subject image as a reference image through an imaging unit, performs focus adjustment when a release operation is detected, and captures and records the subject image under the focus adjustment as a recording image through the imaging unit. There,
Face area specifying means for specifying at least two face areas in the reference image;
The focus adjustment in a plurality of face areas specified by the face area specifying means is sequentially performed for each face area by detecting the release operation once, and the subject image under each focus adjustment is passed through the imaging means. Focus variation continuous shooting means for capturing as a recording image;
An image recording apparatus comprising:   The focus variation continuous shooting unit groups a plurality of face regions specified by the face region specifying unit according to a depth distance by detecting a release operation once, and focuses on a representative face region in each group. 2. The image recording apparatus according to claim 1, wherein the image recording apparatus sequentially adjusts and captures a subject image under each focus adjustment as a recording image through the imaging unit. It further has setting means for setting the number of continuous shootings,
The image recording apparatus according to claim 2, wherein the focus variation continuous shooting unit groups a plurality of face regions specified by the face region specifying unit according to a set number of continuous shootings according to a depth distance.   The focus variation continuous shooting means continuously opens and closes the shutter by detecting a single release operation, and performs focus adjustment according to a predetermined region order for each period during which the shutter is closed. The image recording apparatus according to any one of 1 to 3.   Further, image composing means for composing a focusing area of each recording image with the reference image by using one of the plurality of recording images captured under the respective focus adjustments as a reference image. The image recording apparatus according to claim 1, wherein the image recording apparatus is an image recording apparatus.   5. The image forming apparatus according to claim 1, further comprising: synthesis information embedding means for embedding a focus position of each recording image captured under the respective focus adjustments in the recording image as information at the time of image synthesis. The image recording device according to any one of the above.   A shooting method characterized by automatically switching an autofocus area to each face area recognized as a face by face recognition processing every time shooting is performed in high-speed continuous shooting.   Extracting and synthesizing images of partial areas that have become the autofocus area from a plurality of recording images generated by photographing each face area as an autofocus area in a predetermined order. The imaging method according to claim 7.

Images